User blog:Homuhomu123/Mid-Term Report on the Artillery Spotting TRE Project
This report will summarize all the experiment data so far I collected on the trigger rates of artillery spotting. Source data are mainly from Kuubowokyu (K-Test) & Homuhomu123 (H-Test). According to limited sample sizes, some of the results could only serve as rough estimate on the actual values. In addition, all of the data analysis part below is done by Homuhomu123, not the author of K-Test. Conclusions from each exp't have been summarized and recorded onto the page Artillery Spotting . Unless otherwise mentioned, all the tests below were done under the following conditions: *Map 3-2-A *Using formation Line Ahead *Test ships were kept beyond minor damage *Morale 40~49 *Ship levels: around 90+ *Non-flagship position. Since the probability of occurence are around 50%, sufficient samples are mandatory for making better judgement.Thus sample sizes below 160 will be shown in brackets (). Effects from each factor are tested as follow: Fleet total LoS (FLoS) DA K-Test: No.1 vs No.3 Estimation: *'+9% DA chance' (on avg) per 100 additional FLoS. (BB(V), CAV) K-Test: No.10, 11, 12 *Map: 1-1-1 Estimation: *FLoS (133 ~> 218): +7.4% DA chance per 100 additional LoS *FLoS (218 ~> 286): +8.4% DA chance per 100 additional LoS Homu Test: Exp't K, L, M & N *Red morale test. *For Exp't K, ships were kept beyond moderate damage. *Sample size ~300. Estimation: *For CA(V), their DA chances will change by around 6% per every 100 FLoS.'' (8 data points in total)'' *In specific, CA Myoukou (5%) & CAV Tone (7%). *For BBV, their DA chances change will change by around 10% per every 100 FLoS.'' (2 data points, rough estimate)'' CI K-Test: No.2 vs No.4 Estimation: *'+10% CI chance' per 100 additional fleet total LoS. Shell K-Test: No.7+8 vs. No.19 *AP shell hybrid setup. (2*main + AP shell + seaplane) Estimation: *'+5% CI chance, +0% DA chance,' per 100 additional fleet total LoS. "FLoS & AP Shell Special Attacks" *Map 1-2-1 *'Flagship position' *2 setups are being investigated here, namely: **AP shell Hybrid setup: 2*Main + AP shell, Seaplane **''AP shell Cut-in setup Main + Sec, AP shell, Seaplane.'' *Sample sizes: 200 Estimation: *'"AP Shell Hybrid Setup"' **'+5% Cut-in' chance per every 100 FLoS. **DA chance is''' independent''' of FLoS. (~30%) **Shown below as the 3 bluish curves. *''"AP Shell Cut-in Setup"'' **'+0.7%' AP cut-in chance per 100 FLoS. (2 data, rough estimate) **'+1.4%' Sec cut-in chance per 100 FLoS. (2 data, rough estimate) **Shown below as the 3 reddish curves. Flagship Status DA K-Test: No.7, 8, 9 If we neglect the tiny differences in FLoS, we can combine the data as follow: Estimation: *'+17% DA chance' for the flaghship *Basic DA chance for CAs at FLoS = 250 is ~47%. "Kinugasa Test" *Map 1-1-1. *Health was controlled in either full (>75% max) or minor damage（小破）.. *Always sparkled. *Sample size: 200 Estimation: *In the full health state, Kinugasa's DA chance increased by 15.9% when placing as flagship. (2 data points, rough estimate).. *In the minor damage state, Kinugasa's DA chance increased by 10.5% when placing as flagship. (2 data points, rough estimate).. CI K-Test: No.7, 8, 9 If we neglect the tiny differences in FLoS, we can combine the data as follow:: Estimation: *'+13.7% CI chance' for the flaghship. *Basic CI chance for CAs at FLoS = 250 is ~43%. Shell K-Test: No.7+8 vs. No.9 *AP shell hybrid setup (2*main + AP shell + seaplane) Estimation: *'+7.5% on both CI & DA chances' for the flagship using AP hybrid setup. (2 data, rough estimate) "FLoS & AP Shell Special Attacks" vs "Test on the Adv.Setups" *Map 1-2-1 (both) *Flagship position in the 1st test, Non-flagship position in the 2nd test. *2 setups are being investigated here, namely: **AP shell Hybrid setup: 2*Main + AP shell, Seaplane **''AP shell Cut-in setup: Main + Sec, AP shell, Seaplane.'' *Very close FLoS in the 2 tests, thus highly comparable. *Sample sizes: 200+ / 360+ Estimation: *'"AP Shell Hybrid Setup" vs "2*Main + AP (CI+DA)" :' **'+4% DA' chance for the flagship. (2 data, rough estimate) **'+7% CI' chance for the flagship. (2 data, rough estimate) **2* Main + AP + Seaplane *''"AP Shell Cut-in Setup" vs "Sec + AP (CI)"'' **'+17% AP cut-in' chance for the flagship (2 data, rough estimate) **'''-6.5% Sec cut-in''' chance for the flagship. (2 data, rough estimate) **''Main + Sec + AP + Seaplane'' Air Control State (AS vs. AS+) & CI K-Test: No.17 vs No.19 *Formation: Diamond *Map: 5-4-1 Estimation: *'''-12% DA / CI chance''' if only AS is achieved. HYBRID K-Test: No.22 vs No.23 Estimation: *'''-11% CI chance &''' -7% DA chance for a ship using AP hybrid, if only AS is achieved. (rough estimate) Fourth Slot Equipment (Radar, AP Shell or Extra Seaplane) CI K-Test: No.2, 4, 13 & 16 Estimation: *No.32 (L) Surface boosts CI chance by ~30% *No.22 (S) Surface boosts CI chance by ~28% (2 data, rough estimate) *No.14 boosts CI chance by ~30%.(2 data, rough estimate) *AP shell boosts CI chance by ~20% (2 data, rough estimate) *An extra seaplane doesn't really affect CI chance. (2 data, rough estimate) Homu Test: Exp't G, H & J ' *Red morale test.. *FLoS ~300. *Sample size 170 ~ 200. Estimation: ''(2 data for each, rough estimates) *By using #32 (L) Surface Radar as reference: *FuMO25 (L) gives 4% less CI chance. (+ 26%) *No.21 (L) Anti-Air gives 8% less CI chance. (+ 22%) *No.22 Kai-4 (S) Surface gives 5% less CI chance. (+ 25%) *No.22 (S) Surface gives 6% less CI chance. (+ 24%) DA '''K-Test: No.6a+b, 14 & 15. Estimation: (2 data for each, rough estimates) *No.32 (L) Surface boosts DA chance by ~10% *An extra seaplane may increase DA chance by ~6% (rough estimate) *No.14 (L) Anti-Air boosts DA chance by no more than 6%, (rough estimate).. *No.33 (S) Surface gives ~3% higher DA chance than No.14. (rough estimate). *No.22 (S) Surface boosts DA chance by no more than 4%. Homu Test: Exp't F vs K ''' *Red morale test *Ships were kept beyond moderate damage. *Sample size 250 ~ 300. Estimation: *No.33 (S) Surface boosts DA chance by around '''2% *No.32 (L) Surface boosts DA chance by ~10%. *FuMO (L) increases DA chance by around 6% (rough estimate) "Kinugasa Test" *Map 1-1-1. *Always sparkled. *Flagship position.(Kinugasa K2) *Sample size: 200 Estimation: *An extra Type 0 Obsv.Seaplane increases DA chance by 11.9%.'' (2 data points, rough estimate)'' *No.32 (L) Surface Radar increases DA chance by 13.5%.(2 data points, rough estimate) Type of Seaplane DA K-Test: No.20 Estimation (2 data for each, rough estimates) *Using Type 0 Recon's value as reference *Type 98 Night Scout gives ~2% lower DA chance (rough estimate) *Type 0 Observation Seaplane.& Prototype Seiran gives ~2% higer DA chance. (rough estimate) CI 【艦これ】結局紫雲とかAr196改ってどうなのさ？ (Click here for the spreadsheet) *Formation: Double Line *Sample size 360 ~ 400. Estimation *Different types of seaplane have negligible effect on daytime CI chance. HYBRID K-Test: No.19 Estimation: *Type 0 Observation Seaplane gives ~3% higher DA chance, when using the AP hybrid setup. (rough estimate) Type & Number of Calibur Guns DA K-Test: No.21 Estimation: *Different types of caliber guns doesn't really affect DA chance. (rough estimate) *An additional main gun doesn't really affect DA chance..(rough estimate) CI K-Test: No.19 Estimation: *Additional seconday gun doesn't really affect CI chance. Health State DA Homu Test: Exp't L vs K ''' *Red morale test *Ships were either''' kept beyond minor damage, or in moderate damage.. *Sample size 270 ~ 330. Estimation: *Minor damage: negligible '''? *Moderate damage: '''negligible, Modifer no greater than -5%. (90% confidence) "Kinugasa Test" *Map 1-1-1. *In either full health or minor damage（小破）. *Always sparkled. *Sample size: 200 Estimation: *When placed as flagship, DA chance decreased by 6.9% in the minor damage state (2 data points, rough estimate).. *When placed as non-flagship, DA chance decreased by 1.5% in the minor damage state. (2 data points, rough estimate).. Leveling (Individual LoS + Lv.) DA Homu Test: Exp't M & N ''' *Red morale test *Sample size ~300. Estimation: *Myoukou class CAs: DA chance '''increased by 7.1% from ~Lv.37 to ~Lv.73, but only +0.6% from ~Lv.73 to Lv.96+.'' (2 data each, rough estimates)'' *Nagato class BBs: DA chance decreased by 1.9% from ~Lv.42 to ~Lv.96. (2 data, rough estimate) *Inconclusive. Category:Blog posts